metadata
base_model: mistralai/Mistral-7B-Instruct-v0.3
datasets:
- GaetanMichelet/chat-60_ft_task-3
library_name: peft
license: apache-2.0
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: Mistral-7B_task-3_60-samples_config-1_full
results: []
Mistral-7B_task-3_60-samples_config-1_full
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60_ft_task-3 dataset. It achieves the following results on the evaluation set:
- Loss: 0.8734
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.2252 | 0.8696 | 5 | 1.1981 |
1.1411 | 1.9130 | 11 | 1.0844 |
1.0554 | 2.9565 | 17 | 0.9996 |
0.8972 | 4.0 | 23 | 0.9074 |
0.8159 | 4.8696 | 28 | 0.8847 |
0.7851 | 5.9130 | 34 | 0.8734 |
0.7179 | 6.9565 | 40 | 0.8734 |
0.6768 | 8.0 | 46 | 0.8809 |
0.6127 | 8.8696 | 51 | 0.8960 |
0.5592 | 9.9130 | 57 | 0.9449 |
0.4519 | 10.9565 | 63 | 0.9643 |
0.367 | 12.0 | 69 | 0.9817 |
0.3385 | 12.8696 | 74 | 1.1056 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1